Daniel Kahnrmann

THINKING, FAST AND SLOW

This idea is explored in detail throughout the book, with abundant citing of research, which may suggest it is a dry read. That is far from the case; Kahnemann writes with plenty of humour and is never dull. But he does expect his reader to cooperate by doing some of the work; there are little problems to solve and frequent requests for decisions about what you would do in various imaginary scenarios, many involving gambles. The conclusions that emerge are mostly counter-intuitive, so it is necessary to read fairly slowly to avoid missing the frequent subtleties of the discussion.

The book has five parts. Part 1 is about the differences between System 1 and System 2. Part 2 explores the question of why it is so difficult, even for professional statisticians, to think statistically. Part 3 takes this theme further and considers why we are excessively confident in what we believe we know. "We are prone to overestimate how much we understand about the world and underestimate the role of chance in events."

Part 4 is a little more technical and deals with something called prospect theory, a model of choice in economics that Kahnemann and Amos published in 1979. Kahnemann argues that choices are often less than fully rational, owing to the influence of System 1. The emphasis in this section is more on psychology than on economics per se (Kahnemann is not a professional economist).

Part 5 introduces another binary distinction in how we think, based on recent research which shows the existence of two selves, the remembering self and the experiencing self. Happiness and unhappiness seem to depend more on remembering than on experiencing. And it is remembering that gives rise to our sense of self. "Odd as it may seem, I am my remembering self, and the experiencing self, who does my living, is like a stranger to me."

Two papers by Kahnemann and Tversky are included as appendices.

There are so many original ideas in this book that it seems misleading to pick out any individually, although I'd want to highlight the important discussion of the statistical phenomenon known as regression to the mean. Lack of awareness of this leads to a huge number of false conclusions, notably in medicine. Others that struck me include the unreliability of investment advisers (something I'd long suspected), the fallibility of the interview as a selection tool, and the unsuspected role of chance in the success of business enterprises such as Google.

The ultimate test of an explanation [of why a business succeeds] is whether it would have made the event predictable in advance. No story of Google's unlikely success will meet that test, because no story can include the myriad of events that would have caused a different outcome.

After two years' work had been completed, Kahnemann asked one team member, Seymour, to describe the experience of other teams working on similar projects (something Seymour had already studied). He reported that the chances of success were small—about 40% after seven years' work with no assurance of even completing the task! But the team persevered regardless. A textbook was eventually produced after 8 years, long after Kahnemann's departure, but it was never used. "Facing a choice, we gave up rationality rather than give up the enterprise."

Kahnemann's final assessment of this episode is illuminating—and characteristically honest.

For many years I thought that the main point of the curriculum story was what I had learned about my friend Seymour: that his best guess about the future of our project was not informed by what he knew about similar projects. I came off quite well in my telling of the story, in which I had the role of clever questioner and astute psychologist. I only recently realized that I had actually played the roles of chief dunce and inept leader.